1. Field of the Invention
The present invention relates generally to image processing apparatuses for carrying out a variable scale magnification processing to contract or enlarge an original image and a filtering processing to maintain an image quality of the original image and, more particularly, to an image processing apparatus capable of preventing a deterioration in image quality irrespectively of variable scale magnifications.
2. Description of the Related Art
A digital copier is comprised of a scanning system for reading an original (original image) by fractionizing the same into pixels, an image processing apparatus for subjecting the read image to various image processings, and a printer apparatus for forming a copied image in accordance with image data subjected to the image processings.
An original is scanned normally by shifting a one-dimensional image sensor in a main scanning direction to a subscanning direction.
Major image processings are a tone processing for converting multi-valued data into binary data by a dither method or the like in order to reproduce a half tone, a gamma processing for correcting densities in accordance with visual characteristics, a processing for an image edition such as color changing, a variable scale magnification processing for enlarging and contracting an original image depending on a copying magnification, and the like.
In the variable scale magnification processing, when the original image is contracted, pixel data (which are, strictly speaking, information indicating the shade or the tone of pixels and hereinafter referred to as pixels) are skipped. When the original image is enlarged, pixels are increased. More specifically, in a variable scale magnification processing with the direction of variable scale magnification being a main scanning direction, if a variable scale magnification is, for example, "0.5", every other pixels arranged in the main scanning direction are skipped. If variable scale magnification is, for example, "3", the respective pixels are increased so that the same pixels are arranged three by three in the main scanning direction.
In a series of image reproducing steps such as a scanning of an original image, an image processing and a copying (image formation), an image quality deteriorates due to various factors in accordance with transmission of image information. For example, there occur image blurring due to aberration of an optical lens, Moire fringe (a striped design) due to an interference with a dither pattern, and the like.
Thus, in addition to the foregoing processings, an image quality correction processing for maintaining the image quality of the original image has conventionally been carried in an image processing apparatus. The image quality correction processing can be divided into an input-side correction for restoring an image quality which is already deteriorated at the time when an image (image data) is input to the image processing apparatus, to the one before deteriorated and an output-side correction for adjusting an image quality in advance in view of the deterioration thereof after output from the image processing apparatus.
Normally, a so-called image quality emphasizing processing for intentionally adjusting the image quality in accordance with a subjective preference with respect to a copied image, e.g., an edge emphasis processing and a soft focus processing are carried out along with the image quality correction processing.
In the specification, processings other than the image quality correction processing which are made by the image processing apparatus are called "other processings."
In general, the image quality correction processing is carried out by employing a so-called discrete convolution operation (filter processing) using a digital filter of a predetermined matrix size in which an appropriate weighting matrix is determined for each pixel.
At that time, the input-side correction and the output-side correction can separately be carried out before and after other processings; however, in that case, two digital filters are necessary, so that a circuit configuration of the image processing apparatus becomes complicated. Also, a filter processing accompanied by data delay is carried out twice, so that an output of image data is delayed.
Accordingly, conventionally, a single digital filter in which a weighting matrix is elected is provided at a preceding stage or a succeeding stage of other processings so as to carry out both the input-side correction and the output-side correction at the same time, whereby an image quality correction is made by a single filter processing.
Conventionally, the weighting matrix of a digital filter is elected so that the image quality of an original image may be obtained upon, for example, equal scale magnification copying (variable scale magnification is 1), and its value is set to a fixed value.
Thus, there was a problem that the image quality deteriorates as the value of the variable scale magnification becomes distant from 1 in variable scale magnification copying.
That is to say, when an image quality correction processing is carried out before a variable scale magnification processing, if the weighting matrix of a digital filter is fixed to a value provided at the time of equal scale magnification, then the input-side correction becomes appropriate independently of variable scale magnification, however, the output-side correction becomes inappropriate at the time of variable scale magnification. Conversely, when the image quality correction processing is carried out after the variable scale magnification processing, if the weighting matrix is fixed to a value provided at the time of equal scale magnification, then the input-side correction becomes inappropriate at the time of variable scale magnification.